Connectivity is one of the major concerns in human brain mapping. It shows the connections across different brain regions through the nervous system. Until now, the connectivity between the electroencephalogram (EEG) signals has been calculated without taking into the consideration of volume conduction. Even though some of the methods show the flow across the scalp sources, we need a prior assumption about active brain regions. In this chapter, we suggest a new strategy to identify brain sources with their corresponding locations and amplitudes depending on a particle filter. Modeling of the time series (multivariate autoregressive) is used to detect movement and time dependence among the brain sources. Finally, Granger causality techniques have been applied to assess directional causal flow across the sources. We provide a framework to test the analytical pipeline on real EEG information. The results indicate that the suggested strategy is useful for evaluating the directional connections between EEG neural sources.